Who owns online communities?
First, a case study.
Jeff Harrington, the founder of New New Music has been pushed too far / is having a strop. He’s renamed the site to “New Music Shit Hole” and deleted a lot of the content. He sort of explains why in a post, in which he complains about trolling. The money quote is, “The big picture is that this online new music scene is basically 200 guys in their basements with MIDI synthesizers.”
What’s mostly surprising about this development is the timing, as the community had been surprisingly active lately. Last week, they released a compilation CD which they were selling will all proceeds going to benefit Haiti. Shortly before that, there was some sort of contest in which a great number of one minute pieces were created. The site actually was mostly populated by hobbyists (in their basement with synthesizers), which is why I wandered off. But aesthetically, some of the stuff that was going on there was worth paying attention to.
On the other hand, the number of people in the world who actually have interesting things to say about music is very small. The number of people who have interesting musical thoughts is less small, but putting those thoughts into words is notoriously difficult and can distract attention from putting those thoughts into music. So when composers (or wannabe composers) start talking about stuff, they most often start talking about sort of side issues like technology or gossip about composers or economic issues related to music or some kind of ideological whatever.
Tech talk abounds on the internet. It should be avoided if you actually want to make anything. Hobbyists don’t tend to have any good gossip and their economic interests are sometimes contrary to mine. As for ideology, well, this is probably where the flame wars came from.
So, in frustration, Mr Harrington renamed the site and deleted a bunch of content. There were no other moderators, and possibly no backups. The site, which was clearly valuable enough to have been doing projects even last week, is dead.
Community Projects are Hard
I worked for a couple of years on one for my day job, back when I was a music hobbyist and then, after enough time has passed for me to kind of forget the horror, I spent a year or so helping moderate a high traffic community on live journal. There is practical advice that helps: Have a team of moderators, enough so that if one or two of you wander a way for a bit, things keep going. Replace moderators when they burn out, which will take two years max. Have rules against stupid flamey crap and enforce them. Ban disruptive people – even if somebody is a god of music, it doesn’t mean they can participate well in a community.
One person trying to run a large community site is pretty much a guarantee that it will go up in flames. Which it did. And with it went the content. But any number of other things could have gone wrong. Ning, the company that hosts it, could have inexplicably decided to shut down this particular site (maybe they’re uptown). Or they could have decided to cease operations entirely. These scenarios expose a fundamental flaw in the way that community sites are structured.
A lot of people make content, but a few people own the distribution of it. I’m not talking about copyright, since a lot of the more casual content, like playlists, that populate this kind of site shouldn’t be under copyright anyway. But users spend time creating these and their discussions and comments and this is what actually forms the substance of the community. Yet, a very small number of people – the admins, the moderators, the hosting company – are actually the ones directly in control of the integrity of the content. The power of a community is it’s distributed user base, but all of it ends up concentrated in a single point of weakness, vulnerable to the whims of a few.
Web pages make for very nice front ends, and they can be a great way to organize how content is accessed, but in terms of actually working well with actual people who are not being paid to run them: usenet has a way better model. All the pre-web stuff was better designed for working with communities and had robust implementations. IRC and usenet are distributed. The content lives across many servers. There is no master copy. No one person can destroy a group. And yet groups can still be moderated.
The web was originally exciting because of inline images and some formatting stuff. It looked pretty. This might not sound like much, but back in those days, you couldn’t open a windows-created word file on a macintosh computer because there was nothing in common for different platforms of home users (a situation we are happily skipping back towards with phones, but that’s another post). The web let you actually all look at the same document. And it had pictures!
There’s a lot of reason to love that. We can all have our own little space which is ours and put up pictures of our pets and it was centralized in that we controlled our own space. Corporations loved the control aspect. They could entirely run some service and get users to pay for it and show them ads and if any user becomes annoying, they can be expelled. It’s sort of feudal, but, ooh, pictures! The major trend of web 2.0 is not user-generated content, since the early days of the web were all user generated. The major trend is centralization and corporate control. This transfers ownership of our content to a much smaller groups of people or individuals. They may treat it appropriately, or they might get really tired of trolls and delete all of it.
Everything old is new again in 2010. We need to go back to usenet and use it the basis for how community back ends work. If Net New Music wants to get going again, they should probably reconvene there.